Skip to main content

Learning Where to Look – Generative NAS is Surprisingly Efficient

  • Conference paper
  • First Online:
Computer Vision – ECCV 2022 (ECCV 2022)

Abstract

The efficient, automated search for well-performing neural architectures (NAS) has drawn increasing attention in the recent past. Thereby, the predominant research objective is to reduce the necessity of costly evaluations of neural architectures while efficiently exploring large search spaces. To this aim, surrogate models embed architectures in a latent space and predict their performance, while generative models for neural architectures enable optimization-based search within the latent space the generator draws from. Both, surrogate and generative models, have the aim of facilitating query-efficient search in a well-structured latent space. In this paper, we further improve the trade-off between query-efficiency and promising architecture generation by leveraging advantages from both, efficient surrogate models and generative design. To this end, we propose a generative model, paired with a surrogate predictor, that iteratively learns to generate samples from increasingly promising latent subspaces. This approach leads to very effective and efficient architecture search, while keeping the query amount low. In addition, our approach allows in a straightforward manner to jointly optimize for multiple objectives such as accuracy and hardware latency. We show the benefit of this approach not only w.r.t. the optimization of architectures for highest classification accuracy but also in the context of hardware constraints and outperform state-of-the-art methods on several NAS benchmarks for single and multiple objectives. We also achieve state-of-the-art performance on ImageNet. The code is available at https://github.com/jovitalukasik/AG-Net.

J. Lukasik and S. Jung—Authors contributed equally.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 89.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 119.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    We reran this experiment using the implementation from [47].

  2. 2.

    We reran these experiments using the official implementation from [44,45,46], with the same initial training data and amount of top k architectures as for AG-Net.

References

  1. Bender, G., Kindermans, P., Zoph, B., Vasudevan, V., Le, Q.V.: Understanding and simplifying one-shot architecture search. In: ICML (2018)

    Google Scholar 

  2. Bergstra, J., Bengio, Y.: Random search for hyper-parameter optimization. J. Mach. Learn. Res. 13(10), 281–305 (2012)

    MathSciNet  MATH  Google Scholar 

  3. Cai, H., Zhu, L., Han, S.: ProxylessNAS: direct neural architecture search on target task and hardware. In: ICLR (2019)

    Google Scholar 

  4. Chen, T., Guestrin, C.: XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2016)

    Google Scholar 

  5. Chen, W., Gong, X., Wang, Z.: Neural architecture search on ImageNet in four GPU hours: a theoretically inspired perspective. In: ICLR (2021)

    Google Scholar 

  6. Chen, X., Xie, L., Wu, J., Tian, Q.: Progressive differentiable architecture search: bridging the depth gap between search and evaluation. In: ICCV (2019)

    Google Scholar 

  7. Chrabaszcz, P., Loshchilov, I., Hutter, F.: A downsampled variant of ImageNet as an alternative to the CIFAR datasets. CoRR abs/1707.08819 (2017)

    Google Scholar 

  8. Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: CVPR (2009)

    Google Scholar 

  9. Dong, X., Yang, Y.: One-shot neural architecture search via self-evaluated template network. In: ICCV (2019)

    Google Scholar 

  10. Dong, X., Yang, Y.: Nas-bench-201: extending the scope of reproducible neural architecture search. In: ICLR (2020)

    Google Scholar 

  11. Elsken, T., Metzen, J.H., Hutter, F.: Neural architecture search: a survey. J. Mach. Learn. Res. 20(1), 1997–2017 (2019)

    MathSciNet  MATH  Google Scholar 

  12. Goodfellow, I.J., et al.: Generative adversarial networks. arXiv preprint arXiv:1406.2661 (2014)

  13. He, K., Zhang, X., Ren, S., Sun, J.: Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. In: ICCV (2015)

    Google Scholar 

  14. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)

    Google Scholar 

  15. Huang, S., Chu, W.: Searching by generating: flexible and efficient one-shot NAS with architecture generator. In: CVPR (2021)

    Google Scholar 

  16. Kandasamy, K., Neiswanger, W., Schneider, J., Póczos, B., Xing, E.P.: Neural architecture search with bayesian optimisation and optimal transport. In: NIPS (2018)

    Google Scholar 

  17. Kingma, D.P., Welling, M.: Auto-encoding variational bayes. In: ICLR (2014)

    Google Scholar 

  18. Klyuchnikov, N., Trofimov, I., Artemova, E., Salnikov, M., Fedorov, M., Burnaev, E.: NAS-Bench-NLP: neural architecture search benchmark for natural language processing. CoRR abs/2006.07116 (2020)

    Google Scholar 

  19. Krizhevsky, A.: Learning multiple layers of features from tiny images (2009)

    Google Scholar 

  20. Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: NIPS (2012)

    Google Scholar 

  21. Li, C., et al.: HW-NAS-Bench: hardware-aware neural architecture search benchmark. In: ICLR (2021)

    Google Scholar 

  22. Li, L., Talwalkar, A.: Random search and reproducibility for neural architecture search. In: UAI (2019)

    Google Scholar 

  23. Li, Y., Vinyals, O., Dyer, C., Pascanu, R., Battaglia, P.W.: Learning deep generative models of graphs. CoRR abs/1803.03324 (2018)

    Google Scholar 

  24. Liu, C., et al.: Progressive neural architecture search. In: ECCV (2018)

    Google Scholar 

  25. Liu, H., Simonyan, K., Yang, Y.: DARTS: differentiable architecture search (2019)

    Google Scholar 

  26. Lukasik, J., Friede, D., Zela, A., Hutter, F., Keuper, M.: Smooth variational graph embeddings for efficient neural architecture search. In: IJCNN (2021)

    Google Scholar 

  27. Luo, R., Tian, F., Qin, T., Chen, E., Liu, T.: Neural architecture optimization. In: NeurIPS (2018)

    Google Scholar 

  28. Mehta, Y., et al.: NAS-bench-suite: NAS evaluation is (now) surprisingly easy. CoRR abs/2201.13396 (2022)

    Google Scholar 

  29. Mikolov, T., Karafiát, M., Burget, L., Cernocký, J., Khudanpur, S.: Recurrent neural network based language model. In: Kobayashi, T., Hirose, K., Nakamura, S. (eds.) INTERSPEECH 2010, 11th Annual Conference of the International Speech Communication Association, Makuhari, Chiba, Japan, 26–30 September 2010 (2010)

    Google Scholar 

  30. Pham, H., Guan, M.Y., Zoph, B., Le, Q.V., Dean, J.: Efficient neural architecture search via parameter sharing. In: ICML (2018)

    Google Scholar 

  31. Real, E., Aggarwal, A., Huang, Y., Le, Q.V.: Regularized evolution for image classifier architecture search. In: AAAI (2019)

    Google Scholar 

  32. Real, E., et al.: Large-scale evolution of image classifiers. In: ICML (2017)

    Google Scholar 

  33. Rezaei, S.S.C., et al.: Generative adversarial neural architecture search. In: IJCAI (2021)

    Google Scholar 

  34. Rolínek, M., Musil, V., Paulus, A., P., M.V., Michaelis, C., Martius, G.: Optimizing rank-based metrics with blackbox differentiation. In: CVPR (2020)

    Google Scholar 

  35. Ru, B., Wan, X., Dong, X., Osborne, M.: Interpretable neural architecture search via bayesian optimisation with Weisfeiler-Lehman kernels (2021)

    Google Scholar 

  36. Shi, H., Pi, R., Xu, H., Li, Z., Kwok, J.T., Zhang, T.: Multi-objective neural architecture search via predictive network performance optimization. CoRR abs/1911.09336 (2019)

    Google Scholar 

  37. Siems, J., Zimmer, L., Zela, A., Lukasik, J., Keuper, M., Hutter, F.: Nas-bench-301 and the case for surrogate benchmarks for neural architecture search. CoRR abs/2008.09777 (2020)

    Google Scholar 

  38. Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: ICLR (2014)

    Google Scholar 

  39. Snoek, J., et al.: Scalable bayesian optimization using deep neural networks. In: ICML (2015)

    Google Scholar 

  40. Szegedy, C., et al.: Going deeper with convolutions. In: CVPR (2015)

    Google Scholar 

  41. Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the Inception architecture for Computer Vision. In: CVPR (2016)

    Google Scholar 

  42. Tripp, A., Daxberger, E., Hernández-Lobato, J.M.: Sample-efficient optimization in the latent space of deep generative models via weighted retraining. In: NeurIPS (2020)

    Google Scholar 

  43. Wen, W., Liu, H., Chen, Y., Li, H.H., Bender, G., Kindermans, P.: Neural predictor for neural architecture search. In: ECCV (2020)

    Google Scholar 

  44. White, C., Neiswanger, W., Nolen, S., Savani, Y.: A study on encodings for neural architecture search. In: NeurIPS (2020)

    Google Scholar 

  45. White, C., Neiswanger, W., Savani, Y.: Bananas: bayesian optimization with neural architectures for neural architecture search. In: AAAI (2021)

    Google Scholar 

  46. White, C., Nolen, S., Savani, Y.: Exploring the loss landscape in neural architecture search (2021)

    Google Scholar 

  47. White, C., Zela, A., Ru, B., Liu, Y., Hutter, F.: How powerful are performance predictors in neural architecture search? (2021)

    Google Scholar 

  48. Wu, J., et al.: Stronger NAS with weaker predictors. In: NeurIPS (2021)

    Google Scholar 

  49. Xie, S., Zheng, H., Liu, C., Lin, L.: SNAS: stochastic neural architecture search. In: ICLR (2019)

    Google Scholar 

  50. Xu, K., Hu, W., Leskovec, J., Jegelka, S.: How powerful are graph neural networks? In: ICLR (2019)

    Google Scholar 

  51. Xu, Y., et al.: PC-DARTS: partial channel connections for memory-efficient architecture search. In: ICLR (2020)

    Google Scholar 

  52. Yan, S., White, C., Savani, Y., Hutter, F.: NAS-bench-x11 and the power of learning curves (2021)

    Google Scholar 

  53. Yan, S., Zheng, Y., Ao, W., Zeng, X., Zhang, M.: Does unsupervised architecture representation learning help neural architecture search? In: NeurIPS (2020)

    Google Scholar 

  54. Ying, C., Klein, A., Christiansen, E., Real, E., Murphy, K., Hutter, F.: NAS-bench-101: towards reproducible neural architecture search. In: ICML (2019)

    Google Scholar 

  55. Zela, A., Elsken, T., Saikia, T., Marrakchi, Y., Brox, T., Hutter, F.: Understanding and robustifying differentiable architecture search. In: ICLR (2020)

    Google Scholar 

  56. Zhang, M., Jiang, S., Cui, Z., Garnett, R., Chen, Y.: D-VAE: a variational autoencoder for directed acyclic graphs. In: NIPS (2019)

    Google Scholar 

  57. Zoph, B., Le, Q.V.: Neural architecture search with reinforcement learning. In: ICLR (2017)

    Google Scholar 

  58. Zoph, B., Vasudevan, V., Shlens, J., Le, Q.V.: Learning transferable architectures for scalable image recognition. In: CVPR (2018)

    Google Scholar 

Download references

Acknowledgment

JL and MK acknowledge the German Federal Ministry of Education and Research Foundation via the project DeToL.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Jovita Lukasik .

Editor information

Editors and Affiliations

1 Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 7439 KB)

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Lukasik, J., Jung, S., Keuper, M. (2022). Learning Where to Look – Generative NAS is Surprisingly Efficient. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds) Computer Vision – ECCV 2022. ECCV 2022. Lecture Notes in Computer Science, vol 13683. Springer, Cham. https://doi.org/10.1007/978-3-031-20050-2_16

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-20050-2_16

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-20049-6

  • Online ISBN: 978-3-031-20050-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics